Apply ML with your partners without sharing underlying data

AWS Clean Rooms ML helps you and your partners apply privacy-enhancing controls to safeguard your proprietary data and ML models while generating predictive insights—all without sharing or copying one another’s raw data or models. With AWS Clean Rooms ML custom modeling, you and your partners can bring a custom ML model for training and inference using first-party data and algorithms to apply ML predictions at scale without having to share sensitive intellectual property. You can also use an AWS-authored lookalike model and invite your partners to bring a small sample of their records to a collaboration to generate an expanded set of similar records while protecting your and your partners' underlying data.

Benefits of AWS Clean Rooms ML

Remove the need to share data with your partners to train and deploy ML models

With AWS Clean Rooms ML, your data is only used to train your custom or lookalike model, and your data is not shared among collaborators or used to train AWS models. You can remove your data from Clean Rooms ML or delete a custom model whenever you want, and you can apply privacy-enhancing controls to safeguard sensitive data that you bring to a collaboration.

Bring your own ML model and deploy it with your partners without having share your custom model with them

With AWS Clean Rooms ML custom modeling, you can run ML training and inference using your models, algorithms, and data to generate predictive insights with your partners, without having to share your proprietary models or algorithms that you bring to a collaboration.

Leverage AWS Clean Rooms ML lookalike modeling, an AWS-authored model that can help you improve lookalike segment accuracy by up to 36% compared to industry baselines

With AWS Clean Rooms ML lookalike modeling, you can train a custom, AWS-owned ML model for you and your partners. The AWS-authored model was built and tested across a wide variety of datasets such as news, e-commerce, and streaming video channels. Your data is only used to train your model, data is not shared with either party, and you can remove your data or delete a custom model whenever you want.

Use cases

Optimize advertising campaigns

Advertisers can bring their proprietary model and data into a Clean Rooms collaboration, and invite publishers to join their data to train and deploy a custom ML model that helps them increase campaign effectiveness.

Identify fraudulent financial transactions

Financial institutions can use historical transaction records to train a custom ML model, and invite partners into a Clean Rooms collaboration to detect fraudulent transactions.

Accelerate pharmaceutical research

Research institutions and hospital networks can find candidates who are similar to existing clinical trial participants to accelerate clinical studies.

Determine lookalike segments for marketing

Brands and publishers can model lookalike segments of in-market customers and deliver highly relevant advertising experiences.